SGP 2.5.0.17 hangs collecting darks

This has happened a couple of times now. I have only used it for lights once and that was completely uneventful. But over the last couple of days I have bee building a dark library for the new camera and twice, at random times, I got a windows message that SGP has stopped responding. I haven’t followed the other threads on hanging issues, so I’m not sure if this is related. Looking at the log, the last entry is where SGP logs that the camera has taken too long for the download. Link to log file:

https://www.dropbox.com/s/p5no11w31l1dm7h/sg_logfile_20160317214952.txt?dl=0

Thanks.

Tim

@spokeshave

This is the same issue we have been tracking for a while now. We have definitely reduced its prominence, but it’s still out there… mysteriously… for some users, for some systems. It’s a maddening issue and only seems to apply to “fast-ish” back to back images. Will need to do some more thinking on this…

Thanks, Ken. I thought it seemed similar to problems other people have been having. Interestingly, the last time it happened to me, it was during collection of 600-second darks - so not really “fast-ish” - and that was after banging away on darks for over 17-hours straight without any problems. Weird.

Tim

This is happening to me with some regularity. If there is anything I can do to help diagnose, please let me know.

Tim

Unfortunately there is not. I know exactly what the problem is… just not why it happens or how to resolve it. I also do not know why some people see it once in a blue moon, why some people see it regularly and some people will never see it.

OK… So I have overhauled the way in which SGPro presents preview images. I have never been able to produce this issue in the first place so the new pattern we are using appears to be working fine, but I have no idea if it will resolve the issue, make it worse, or if it will simply just act the same.

Please use 2.5.0.20 to make reports on this (if you are afflicted with this particular issue)

Just a data point - I downloaded 2.5.0.20 today and immediately started taking a series of 300 bias frames. It completed the series without issue. Of course, this is not a definitive test since the problem for me was sporadic, but I would typically get at least one hang on 2.5.0.17 for such a long string of short exposures. So for me, so far, so good.

Tim

I ran another sequence of 500 flats and had no hangs. Still not a certainty, but I think it is much improved.

Tim

Thanks for the report. This new method is faster, recommended by Microsoft and seems more stable (in general). I know it doesn’t work for everyone (as is evidenced by recent reports from @jmacon), but I think we will stay with this method.

2.5.0.21 uses some new code (not super thrilled about it) that is hopefully smart enough to figure out which method to use. Hoping it will work OK for just about everyone.